Hugging Face's logo

nvidia
/
DeepSeek-V4-Flash-NVFP4

Text Generation
Safetensors
Model Optimizer
deepseek_v4
nvidia
ModelOpt
DeepSeekV4
quantized
NVFP4
nvfp4
8-bit precision
fp8
Model card Files
xet
 Community
DeepSeek-V4-Flash-NVFP4
168 GB
  • 1 contributor
 History: 7 commits
jingyux-nv's picture
jingyux-nv
Fix benchmark note: remove inaccurate FP8 baseline line, correct max tokens to 384000
48bfe38 verified